12/2 cooling system, round 2

[Doug]

high concentrations fo clorine will even rust stainless steel ( the reason Idophore is used to disinfect dairies ). I don't know the chemestry behind it.

No electrical equipment should be used around pools and spas without GFI protection.

Doug

[snail]

[Gotta love this forum, ancient history to boot./quote]

It may be ancient history to you, I suspect many members of this forum were there!

Brian

[cujet]

Before you settle on thermosiphon, consider that a small Taco or Grundfos pump will do a great job at circulation with minimal power consumption. Some are as low as 1/25th HP.

This brings distinct advantages. The volume can be reduced to a reasonable level, marine heat exhangers can be used, and the big one, temp diffenential between cyl inlet and head outlet can be reduced to just a few degrees. This is a very good thing when you want to have control of actual internal engine temps, which you seem to need. It would also allow the use of a fluid other than water, if you want to avoid iinternal engine corrosion and risk of boiling.

We found that radiator size can be cut by more than half when using a water pump, vs. thermosiphon.

Remember that your heat exchange is much more effective if the entire surface of the exchanger is a given temp, instead of a temp gradient.

By the way, I think it is truly "worth it". Some good engineering can improve your systems overall efficiency by large numbers!

Chris

[skeeter]

cujet - Thanks for the response. I appreciate a different point of view on this. Maybe I got stuck on idea, an failed to consider even better alternatives. I guess I am bias towards a thermosiphon system because of its simplicity. My reasoning was, that without pumps in the primary (engine to heatx) and secondary (heatx to insulated tank) loops, if properly setup and bled, and coolant levels maintain, their was hardly anything that could go wrong except for a pump failure in the third (getting rid of the transferred heat in the insulated tank) loop. Because of the tanks relatively large volume, their would be some time to recognize the failure and shut it down, before catatrophic engine failur occurred. The "risks" that concerned me most in this system were, insufficient heat exchange capacity, and insufficient engine input/ output temp delta with available flow. Since the stock (although ported) coolant manifolds are being used,  I guess I'm assuming that these will not severly restrict flow (George's CD mentioned this as a possible pitfall). I believe the heatx I have, does have sufficient capacity. Enough of my idea, lets explore your.

I agree, small, efficient pumps are available. I also agree that the volume in the secondary loop could be significantly reduced, lowering the cost of anti-freezing it. By other type fluid in primary loop, are you thinking oil? Pumps would increase heatx efficiency. Can your suggestion be described as follows:

Primary loop - pump 1 output connected to cylinder coolant manifold. 195 deg thermostats on each cylinder. Output of head coolant manifold connected to top port of heatx (circuit 1), bottom port of heatx (circuit 1) connected to pump input. A reservour tank is teed into loop and positioned at a higher level to maintain fluid level.Primary loop contains fluid to reduce engine corosion.

Secondary loop -  pump 2 output connects to heatx bottom port (circuit 2), heatex top port (circuit 2) connects to my system tee/ gate valves (house baseboards, water/ air into garage, pool), cooler returning fluid to pump 2 input. A reservour tank is teed into loop and positioned at a higher level to maintain fluid level.

[skeeter]

You of course want to maintain opposing flows thru the heatex to maintain an even delta across the plates for a steady transfer.

Oh no, their may be a problem with my system.  Although I don't know if its fatal.  After rethinking this, with thermosiphoning , I don't think I can satisfy the above requirement. In my sketch, hot water from the primary loop (engine), enters the heatx at the top port (circuit 1), and cool fluid from the secondary loop (tank), enters the heatx from the bottom port (circuit 2). This would seem to allow a large temp gradient across heatx, lowering effficiency. To do it optimally, I think I would need to raise the tank considerable. Tough with 7 1/2 ft ceilings.

[hotater]

cujet, I'm confused.   

How can a water pump add to the effeciency of the system?   

If the engine has a thermostat, the t'stat *allows* circulation no matter if it's a pump thats been sitting there running against it for an hour, or if it's just hot water rising, when the fluid temp gets to 195. Mine actually opens at 197 in a sauce pan.  The t'stat only stays open long enough to sense cooler water, then mostly or totally closes, depending on temperature.  Mine stays open just enough to circulate at 195, but takes 197 to open it from closed.

  So, how does a pump 'help' except to cut the size of the system down to the point it could cause damage if the pump fails or the weather gets too hot or a plastic bag blows over the radiator....?

I don't see the sense in complicating a Listeroid past installing a t'stat which allows faster heating of the engine on start-up and a constant head temperature afterwards.

To heat a large tank of water by Lister all it takes is a tube type heat exchanger with the ends cut out so the tank water will naturally thermosiphon past the coolant tubes that are just suspended vertically inside the tank, or any combination of small heater cores or radiators that do the same thing.     The only risk would be a leak in the engine coolant that polluted the larger tank, or the larger tank running dry so the exchanger and engine overheat.  The latter could easily be overcome with a simple float cut-off switch.

[skeeter]

Never thought of using a shell and tube inside the tank. I was going to use a coil of  7/8 or 1" or copper tubing,  or submerge a radiator inside the tank if I could find one that would fit. I guess I moved away from that idea when trying to come up with a plan to actually do it.  When standing next to my 80 gal tank, it started to seem like alot more trouble. Trying to figure out how I was going to actually "reach" the ports (near top and bottom), to make the bulkhead connections for the heat exchanger. I also started doubting my plan to adapt the existing ports to leak free bulk head fittings. Sitting here now, I must admit, maybe I shelved the idea too quickly, maybe their is an easy way to accomplish this. When standing in front of the tank, the job seemed a bigger pain in the butt then it does right now. I also must admit, I've really complicated this.

[Tugger]

My tank is 16 inches in diameter by 48 inches high....
My 12/2 will start boiling the water 3.5 hours under heavy loads...
I have 22 feet of 1/2 inch copper tubing running through the tank...
I run 60 degree water in the bottom of the tank and out the top.
It lowers the tank temp by 10 degrees in 2 minutes..
This fall i will try heating my infloor hydronic heat system using the copper tubing as the heat source...and post the results..
Cheers

[spike]

Oh no, their may be a problem with my system.  Although I don't know if its fatal.  After rethinking this, with thermosiphoning , I don't think I can satisfy the above requirement. In my sketch, hot water from the primary loop (engine), enters the heatx at the top port (circuit 1), and cool fluid from the secondary loop (tank), enters the heatx from the bottom port (circuit 2). This would seem to allow a large temp gradient across heatx, lowering effficiency.

You don't have a problem. The way you are suggesting to plumb your heat exchanger, counter flow,  is the recommended and most efficienent. The larger the temperature gradiant between fluids the more heat flow you have and thus, the higher the heat transfer efficiency.

Tim

[skeeter]

Spike - Your right, I had a brain laps. Chalk it up as second guessing myself. Just want to do this once, and get off and running again. I've had this thing idle for too long now.

Finished checking everything out and will reassemble engine tonight. Overall, I have no complaints with what I saw. After 600 hours, the piston big end bearings look the same as they did when I first inspected them.  The rings measured almost no wear and completely free, with clean grooves. Did notice some polishing at very top of pistons.  The cylinders still have well defined crosshatching, no verticle lines, and a "slightly" more polished look, where the top and bottom rings change direction in the cylinder. The only thing I did note with the cylinders before cleaning, was a slight build up at the top of cylinders, making its way a quarter to half the distance to the first ring. This is most likely baked on wvo residue, and was a pain to remove. Probably responsible for the piston polishing noted above. The valves and seats did have some carbon buildup. The crankcase was free of any grit. And as mentioned earlier, oil looked very clean also. I think if I get cooled properly, and with the wvo properly heated now, it should be able to put up some good hours. I plan to pull head after another 500 hour or so to compare.

[ronmar]

You of course want to maintain opposing flows thru the heatex to maintain an even delta across the plates for a steady transfer.

Oh no, their may be a problem with my system.  Although I don't know if its fatal.  After rethinking this, with thermosiphoning , I don't think I can satisfy the above requirement. In my sketch, hot water from the primary loop (engine), enters the heatx at the top port (circuit 1), and cool fluid from the secondary loop (tank), enters the heatx from the bottom port (circuit 2). This would seem to allow a large temp gradient across heatx, lowering effficiency. To do it optimally, I think I would need to raise the tank considerable. Tough with 7 1/2 ft ceilings.

Your statement above just described opposing flows...  Hot water exits the head and travels up to the top of the heatex at say 195F.  As it is exposed to the plates cooled by the secondary loop and gives up it's heat, it cools and falls to the bottom port for the primary loop and exits at say 120F back to the engine cylinder base port.  The cooler water of the secondary loop enters at the bottom of the heatex at say 90F.  As it encounters the plates warmed by the primary loop exit water(120F-90F=30F delta), it absorbs the heat and becomes less dense and rises toward the top of the heatex.  As it rises(and warms), it encounters progressively warmer plate the closer it gets to the 195F primary loop inlet, and continues to absorb heat at a steady rate.  If it exits at 165F it has the same 30 degree delta that it encountered when it started at the bottom and in theory you would also experience a 30F delta along the plates in the middle of the heatex.  

If the secondary loop entered at the same end as the primary loop, you would have a huge delta(195F-90F =105F delta) at the start, but as the two fluids moved along the plates and the primary loop gave up it's heat to the secondary you would have a steadly decreasing delta untill the two fluid temps reached equilibrium.  This would mean a higher primary loop outlet temp and a lower secondary loop outlet say both around 140F.
These of course are just numbers I pulled out of the air to show the relationships.

How much thermosiphon height you need depends on flow restrictions.  Thermosiphon is simply convection.  How does the heating element on a hot water tank get the heat to the top outlet on the tank?  It convects it there(and conducts a little too)...  The thermosiphon works the same way, the tanks could be beside the engine as long as there a fluid path above the heat source and there is a temperature difference or gradient between the top and bottom ports on the tank.  Your real pump force is comming from the engine(or heatex) anyway as it is transfering the heat to the water and causing the lowering of the density.  Thermosiphon works in a circle, just like convection.

Here is a science experiment for you.  Get a tall pan and fill it with water.  Preferably not one of your wifes as you are going to take a torch to it:)  An old metal coffee can works well for this.  Preferably the water will be a little dirty with some floating debris to help you see movement(sprinkle in some pepper).  Take a propane torch and apply heat to one spot about half way up the side(simulates heat input from the engine).  You will almost immediatly see the water on that side begin to rise and spread across the top of the pan.  To enhance the effect hold a chunk of ice against the opposite side of the pan across from the torch(simulates heat removed by the heatex or radiator) which will cool the water along that side and cause it to fall.  If some of the water debris can settle to the bottom, you will see a circular flow establish, up on the heat side and down on the cool side.  Now imagine the same pan with a pipe alongside connected at top and bottom with elbows.  Apply the heat to the pipe and water will flow out the top into the tank(as long as the pipe is below the water level) and cool water will flow in to the bottom  of the pipe.  It dosn't mattter where you apply the heat along the pipe, the effect is the same.  The same thing happens when you place a small pipe over a candle.  The warm air has no where to go but up the chimney, only to draw in fresh cold air at the bottom to feed the cycle.  

[cujet]

Hotater said ""cujet, I'm confused.  How can a water pump add to the effeciency of the system?""

Good Morning Hotater,

I was trying to be careful with words when I wrote that ""Remember that your heat exchange is much more effective if the entire surface of the exchanger is a given temp, instead of a temp gradient.""

I don't want to imply that efficiency will be improved (because that may not be the case), only effectiveness of a given heat exchanger.

I suppose all I was trying to say was that using small (very reliable) pumps in such a system will reduce the overall size of the system and it will allow more precise temperature control of the engine.

Being a motorhead, (ex race car team member, current aircraft mechanic)  I have a good deal of experience with enignes and temperature management. I prefer to run most of my engines at around 210F and no higher due to the nature of coolant. This results in better lifespan. Look up some data on cast iron cylinders and lifespan vs. temp. The real trick is to increase the inlet water temp to the point where the cylinder bore is above 160F. Not much heat is transfered from the lower area of the bore area. Yet, there is often a large water jacket here (often unnecessary).

A properly sized cooling system will provide adequate cooling while creating a minimal Delta T across the engine's inlet/outlet. To do this requires decent water flow velocity. Truly the opposite of a thermosiphon system, which, by nature, has a large Delta T.

Thermosiphon systems can be excellent and truly have a place in the Listeroid world. I currently have mine set up so if my pump fails, thermosiphon will provide the flow.

Chris

[skeeter]

Just a follow-up update of my coolant system upgrade. First, I deviated from my original plan. I discarded the idea of using a large water tank. The system is now configured as follows. From the engine, I'm thermosyphoning to a big flat plate heat exchanger. Total coolant capacity in primary loop is about 6 gals, 5gal in engine and heatx, 1 gal in reservior tank. The secondary loop counterflows with primary, and consists of a taco pump, and 3 valve selectable branches. One branch feeds a water- air heatx, which exhausts the heated air into a soon to be work shop, is now a storage area for wvo. On the hottest day so far (93 deg. F ambient, 108 deg. F room temp), load 75% capacity, this loop was able to maintain outlet coolant temp to no higher then 204 deg. F, with secondary loop outflow from heatex at around 175 deg. F. When this loop is not selected I can optionally keep the fan running, to exhaust hot air from the room. As originally planned, the second branch of the secondary loop, feeds a 30+ feet, 3/4" copper tubing coil submerged in my 24 ft round, 54" deep pool. Pool is 14' from the wall of my engine room. Closed cell form insulated PEX, trenched in 2ft, connects the coil to the engine room plumbing. Water temps in this branch so far, have not exceed 140 deg. on a hot day, and primary coolant exits engine temp varies between 192 & 202 deg. F, as t-stat cycles. So far, I have not connected third branch too house hot water baseboard string. Plan on installing overtemp fuel cutoff switch this week, which I feel is absolutely needed due to coolant system configuration.

[skeeter]

Jens,

This is a big flat plate heatx, 24" x 9 1/2" x 8 1/2", 2 1/2 ports reduced to 1 1/4", 76 plates, plate area of 114 sq ft, weight - 114 lbs. I wouldn't attempt this on a garden variaty heatx. I be sure to post pictures on coppermine, today or tomorrow.

[skeeter]

Photos have been posted on coppermine.